There is increasing consensus that outcomes for substance abuse treatment can be enhanced and extended with the addition of psychosocial approach. The disappointing record of naltrexone, an excellent, long acting opioid antagonist, in the substance abuse treatment system, may in part be due to failure to combine it with psychosocial approaches to enhance compliance with naltexone and reward abstinence. The multifaceted CRA approach developed by Higgins and colleagues is particularly attractive as a strategy to improve outcomes in naltraxone treatment because several of its components could be targeted to directly address naltrexone's weakness. For example, to compensate for naltrexone's lack of pharmacological reward, contingency management through use of a voucher system could be used to reward and enhance naltrexone ingestion. Similarly, involvement of significant others in treatment through reciprocal relationship counseling could be used to provide increased incentives for retention and encouragement to persist with treatment despite protracted withdrawal. Using a dismantling strategy, we will evaluate the value of adding these two distinct components of Higgins' treatment package to standard naltrexone maintenance. 180 recently detoxified opioid addicts will be randomly assigned to either (1) contingency management plus significant other involvement in addition to standard naltrexone treatment, (2) contingency management in addition to standard naltrexone treatment, or (3) standard naltrexone treatment alone, consisting of thrice weekly clinic visits for naltrexone, thrice weekly supervised urine toxicology screens, and weekly group counseling. Primary outcomes will be retention, compliance with naltraxone treatment, and reduction in opioid, cocaine, and other substance use. All treatments will be manual-guided and delivered by experienced clinicians, who will receive training and ongoing supervision to prevent drift during the course of the study. Supervised urines will be collected thrice weekly. All counseling sessions will be videotaped for ongoing therapist supervision and process assessment. Follow-ups conducted 3, 6, 9 and 12 months after the cessation of the contingency management and relationship counseling will assess the durability of treatment effects.